Reducing wind energy forecast uncertainty through improved measurements and modeling

Wind energy is one of the most mature renewable energy technologies and thus has the potential for near-term impact on reducing fossil-fuel dependence. A significant problem with the integration of wind energy onto the electrical grids is that wind is a variable resource that is difficult to predict to the required precision. Much of this forecasting difficulty can be traced to a lack of measurement information in the atmospheric layer occupied by the turbine blades, which lies above the region conveniently measured by instrumented towers. Flows in the turbine-blade layer aloft can often be decoupled from the surface, so that surface-based tower measurements are often not indicative of wind characteristics in this layer. The deficiency of measurement data means that flow phenomena in the blade layer are poorly characterized or understood, forecast accuracy is difficult to evaluate, and NWP performance in this layer is largely unverified, although available studies indicate the need for significant improvements in the models.

The lack of measurement data in the blade layer does not mean that none exists. Besides a few tall towers that have long-term records at isolated sites, some high-quality measurements of flow aloft are available through short-term field programs. Field-project datasets sampling this layer have provided insight into the types of flow phenomena occurring in key geographic areas of importance to wind energy. In this study we review the structure and evolution of several types of flow from Doppler-lidar studies in the U.S. Great Plains and in complex terrain. For more than 20 years NOAA /ESRL has been performing Doppler-lidar studies of complex-terrain flows focusing on the lowest few hundred meters of the atmosphere, and for nearly a decade, pursuing studies of the Great Plains low-level jet and its relationship to the stable boundary layer and to wind energy. Based on these findings we explore the kinds of enhancement to measurement and NWP capabilities that would be required to improve the accuracy of predictions to the degree required to make wind energy a grid-friendly contributor to the national energy portfolio.